Unit root tests with smooth breaks: an application to the Nelson–Plosser data set

Abstract

This article reconsiders the nature of the trends (i.e. deterministic or stochastic) of the Nelson-Plosser macroeconomic time series. For this purpose, the article employs two new tests that display robustness to structural breaks of unknown forms, irrespective of the date and/or location of the breaks. These tests approximate structural changes as smooth processes via flexible Fourier transforms. In general, it appears that real variables are stationary while nominal ones have a unit root.

Acknowledgements

I am grateful to Walter Enders and Junsoo Lee for helpful comments and suggestions. Any remaining errors are my own.